Re: [PATCH 1/1] [PATCH v3]sched/pelt: Fix the attach_entity_load_avg calculate method

[Peter Zijlstra]

I've taken the liberty of carrying over the tags from v2 and reworked
the Changelog a little.

---
Subject: sched/pelt: Fix attach_entity_load_avg() corner case
From: kuyo chang <kuyo.chang@xxxxxxxxxxxx>
Date: Thu, 14 Apr 2022 17:02:20 +0800

From: kuyo chang <kuyo.chang@xxxxxxxxxxxx>

The warning in cfs_rq_is_decayed() triggered:

    SCHED_WARN_ON(cfs_rq->avg.load_avg ||
		  cfs_rq->avg.util_avg ||
		  cfs_rq->avg.runnable_avg)

There exists a corner case in attach_entity_load_avg() which will
cause load_sum to be zero while load_avg will not be.

Consider se_weight is 88761 as per the sched_prio_to_weight[] table.
Further assume the get_pelt_divider() is 47742, this gives:
se->avg.load_avg is 1.

However, calculating load_sum results in 0:

  se->avg.load_sum = div_u64(se->avg.load_avg * se->avg.load_sum, se_weight(se));
  se->avg.load_sum = 1*47742/88761 = 0.

Then enqueue_load_avg() adds this to the cfs_rq totals:

  cfs_rq->avg.load_avg += se->avg.load_avg;
  cfs_rq->avg.load_sum += se_weight(se) * se->avg.load_sum;

Resulting in load_avg being 1 with load_sum is 0, which will trigger
the WARN.

Fixes: f207934fb79d ("sched/fair: Align PELT windows between cfs_rq and its se")
Signed-off-by: kuyo chang <kuyo.chang@xxxxxxxxxxxx>
[peterz: massage changelog]
Signed-off-by: Peter Zijlstra (Intel) <peterz@xxxxxxxxxxxxx>
Reviewed-by: Vincent Guittot <vincent.guittot@xxxxxxxxxx>
Tested-by: Dietmar Eggemann <dietmar.eggemann@xxxxxxx>
Link: https://lkml.kernel.org/r/20220414090229.342-1-kuyo.chang@xxxxxxxxxxxx
---
 kernel/sched/fair.c |   10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -3829,11 +3829,11 @@ static void attach_entity_load_avg(struc
 
 	se->avg.runnable_sum = se->avg.runnable_avg * divider;
 
-	se->avg.load_sum = divider;
-	if (se_weight(se)) {
-		se->avg.load_sum =
-			div_u64(se->avg.load_avg * se->avg.load_sum, se_weight(se));
-	}
+	se->avg.load_sum = se->avg.load_avg * divider;
+	if (se_weight(se) < se->avg.load_sum)
+		se->avg.load_sum = div_u64(se->avg.load_sum, se_weight(se));
+	else
+		se->avg.load_sum = 1;
 
 	enqueue_load_avg(cfs_rq, se);
 	cfs_rq->avg.util_avg += se->avg.util_avg;